Convolutional Neural Networks

Project: Write an Algorithm for a Dog Identification App


In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!

Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.

In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.

Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.

The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.


Why We're Here

In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

Sample Dog Output

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!

The Road Ahead

We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.

  • Step 0: Import Datasets
  • Step 1: Detect Humans
  • Step 2: Detect Dogs
  • Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
  • Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)
  • Step 5: Write your Algorithm
  • Step 6: Test Your Algorithm

Step 0: Import Datasets

Make sure that you've downloaded the required human and dog datasets:

Note: if you are using the Udacity workspace, you DO NOT need to re-download these - they can be found in the /data folder as noted in the cell below.

  • Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dog_images.

  • Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.

Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.

In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.

In [1]:
import numpy as np
from glob import glob

# load filenames for human and dog images
human_files = np.array(glob("/data/lfw/*/*"))
dog_files = np.array(glob("/data/dog_images/*/*/*"))

# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 13233 total human images.
There are 8351 total dog images.

Step 1: Detect Humans

In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.

OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.

In [2]:
import cv2                
import matplotlib.pyplot as plt                        
%matplotlib inline                               

# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')

# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# find faces in image
faces = face_cascade.detectMultiScale(gray)

# print number of faces detected in the image
print('Number of faces detected:', len(faces))

# get bounding box for each detected face
for (x,y,w,h) in faces:
    # add bounding box to color image
    cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
    
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Number of faces detected: 1

Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.

In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.

Write a Human Face Detector

We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.

In [3]:
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
    img = cv2.imread(img_path)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    faces = face_cascade.detectMultiScale(gray)
    return len(faces) > 0

(IMPLEMENTATION) Assess the Human Face Detector

Question 1: Use the code cell below to test the performance of the face_detector function.

  • What percentage of the first 100 images in human_files have a detected human face?
  • What percentage of the first 100 images in dog_files have a detected human face?

Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.

Answer: 2% false negative with human faces 17% false positive with dog faces

In [4]:
from tqdm import tqdm

human_files_short = human_files[:100]
dog_files_short = dog_files[:100]

#-#-# Do NOT modify the code above this line. #-#-#

## TODO: Test the performance of the face_detector algorithm 
## on the images in human_files_short and dog_files_short.
In [5]:
def positive_faces(fn, files):
    res = 0
    for img_file in tqdm(files):
        positive = fn(img_file)
        if positive:
            res += 1
    return res

def test_positive_faces(fn):
    return positive_faces(fn, human_files_short), positive_faces(fn, dog_files_short)
In [6]:
test_positive_faces(face_detector)
100%|██████████| 100/100 [00:02<00:00, 36.26it/s]
100%|██████████| 100/100 [00:29<00:00,  7.33it/s]
Out[6]:
(98, 17)

We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [7]:
### (Optional) 
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.

Step 2: Detect Dogs

In this section, we use a pre-trained model to detect dogs in images.

Obtain Pre-trained VGG-16 Model

The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.

In [8]:
import torch
import torchvision.models as models

# check if CUDA is available
use_cuda = torch.cuda.is_available()
In [9]:
# define VGG16 model
VGG16 = models.vgg16(pretrained=True)

# move model to GPU if CUDA is available
if use_cuda:
    VGG16 = VGG16.cuda()

Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.

(IMPLEMENTATION) Making Predictions with a Pre-trained Model

In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.

Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.

In [10]:
from PIL import Image, ImageFile
import torchvision.transforms as transforms

device = torch.device("cuda")
vgg_img_size = 244, 244

transform = transforms.Compose([
                        transforms.Resize(size=vgg_img_size), #VGG expected size
                        transforms.ToTensor()])

def preprocessed_image(image_name):
    image = Image.open(image_name)
    image = image.convert('RGB')
    image = transform(image).unsqueeze(0)
    return image.to(device, torch.float)
In [11]:
from PIL import Image
import torchvision.transforms as transforms

VGG16 = models.vgg16(pretrained=True)
VGG16 = VGG16.cuda()

def VGG16_predict(img_path):
    '''
    Use pre-trained VGG-16 model to obtain index corresponding to 
    predicted ImageNet class for image at specified path
    
    Args:
        img_path: path to an image
        
    Returns:
        Index corresponding to VGG-16 model's prediction
    '''
    
    ## TODO: Complete the function.
    ## Load and pre-process an image from the given img_path
    ## Return the *index* of the predicted class for that image
    
    img = preprocessed_image(img_path)
     
    res = VGG16(img)
    return torch.max(res,1)[1].item() # predicted class index
In [12]:
VGG16_predict(human_files[0])
Out[12]:
678

(IMPLEMENTATION) Write a Dog Detector

While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).

Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).

In [13]:
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
    ## TODO: Complete the function.
    klass_idx = VGG16_predict(img_path)
    return klass_idx >= 151 and klass_idx <= 268

(IMPLEMENTATION) Assess the Dog Detector

Question 2: Use the code cell below to test the performance of your dog_detector function.

  • What percentage of the images in human_files_short have a detected dog?
  • What percentage of the images in dog_files_short have a detected dog?

Answer:

In [14]:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.

test_positive_faces(dog_detector)
100%|██████████| 100/100 [00:04<00:00, 23.89it/s]
100%|██████████| 100/100 [00:05<00:00, 18.75it/s]
Out[14]:
(0, 97)

We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [15]:
### (Optional) 
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.

Step 3: Create a CNN to Classify Dog Breeds (from Scratch)

Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.

We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.

Brittany Welsh Springer Spaniel

It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).

Curly-Coated Retriever American Water Spaniel

Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.

Yellow Labrador Chocolate Labrador Black Labrador

We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.

Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dog_images/train, dog_images/valid, and dog_images/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!

In [16]:
import os
from torchvision import datasets
from torch.utils.data import DataLoader
import torchvision.transforms as transforms


### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes

normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                 std=[0.229, 0.224, 0.225])

train_transform = transforms.Compose([transforms.Resize(size=244),
                                       transforms.RandomHorizontalFlip(),
                                       transforms.RandomRotation(10),
                                       transforms.CenterCrop(224),
                                       transforms.ToTensor(),
                                       normalize])

valid_transform = transforms.Compose([transforms.Resize(size=244),
                                           transforms.CenterCrop(224),
                                           transforms.ToTensor(),
                                           normalize])

test_transform = transforms.Compose([transforms.Resize(size=244),
                                           transforms.CenterCrop(224),
                                           transforms.ToTensor(),
                                           normalize])

train_data = datasets.ImageFolder('/data/dog_images/train', transform=train_transform)
valid_data = datasets.ImageFolder('/data/dog_images/valid', transform=valid_transform)
test_data = datasets.ImageFolder('/data/dog_images/test', transform=test_transform)


loaders_scratch = {
    'train': DataLoader(train_data, shuffle=True, num_workers=0, batch_size=128),
    'valid': DataLoader(valid_data, shuffle=True, num_workers=0, batch_size=238),
    'test': DataLoader(test_data, shuffle=True, num_workers=0, batch_size=256)
}

Question 3: Describe your chosen procedure for preprocessing the data.

  • How does your code resize the images (by cropping, stretching, etc)? What size did you pick for the input tensor, and why?
  • Did you decide to augment the dataset? If so, how (through translations, flips, rotations, etc)? If not, why not?

Answer: -resized to 244x244, it's small so faster processing, yet was enough for VGG model to deduce image class -trnasforming images randomly without distorting them; that is; scale, rotation and flipping, without blacking out parts, sheer or blackenning out parts of it, will still keep the original data, but prevent the model from catching an unwanted recurrent feature like orientation in images

(IMPLEMENTATION) Model Architecture

Create a CNN to classify dog breed. Use the template in the code cell below.

In [17]:
dog_klass_count = 133
In [18]:
import torch.nn as nn
import torch.nn.functional as F

# define the CNN architecture
class Net(nn.Module):
    ### TODO: choose an architecture, and complete the class
    def __init__(self):
        super(Net, self).__init__()
        ## Define layers of a CNN
        

        self.conv = nn.ModuleList([
            nn.Conv2d(3, 16, 3, padding=1),
            nn.Conv2d(16, 32, 3, padding=1),
            nn.Conv2d(32, 64, 3, padding=1),
            nn.Conv2d(64, 128, 3, padding=1),
            nn.Conv2d(128, 256, 3, padding=1),
        ])
        
        self.pool=nn.MaxPool2d(2, 2)
        
        self.dropout = nn.Dropout(0.5)
                        
        self.fc = nn.ModuleList([
            nn.Linear(7 * 7 * 256, 512),
            nn.Linear(512, 256),
            nn.Linear(256, dog_klass_count)
        ])
        
        self.dropout = nn.Dropout(0.5)
    
    def forward(self, x):
        ## Define forward behavio
    
        for layer in self.conv:
            #print(layer)
            x = self.pool(F.relu(layer(x), 3))
           
        x = x.view(-1, 7 * 7 * 256)
        
        for layer in self.fc[:-1]:
            #print(layer)
            x = F.relu(layer(x))
            x = self.dropout(x)

        self.fc[-1](x)
        
        return x

#-#-# You so NOT have to modify the code below this line. #-#-#

# instantiate the CNN
model_scratch = Net()

# move tensors to GPU if CUDA is available
if use_cuda:
    model_scratch = model_scratch.cuda()

Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.

Answer:

I used an architecture like that suggested in classroom videos and similar to that of vgg16 model

I tried 3 layers convolutional layers, but its best achieved accuracy was about 8%, so I increased to 5 layers.

for classifier I increased count of layers to three fully conncted layers, output channel of the last layer is the count of target dog breeds

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.

In [19]:
import torch.optim as optim

### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()

### TODO: select optimizer
optimizer_scratch = optim.Adam(model_scratch.parameters(), lr=0.0001)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.

In [20]:
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
    """returns trained model"""
    # initialize tracker for minimum validation loss
    valid_loss_min = np.Inf 
    
    for epoch in range(1, n_epochs+1):
        # initialize variables to monitor training and validation loss
        train_loss = 0.0
        valid_loss = 0.0
        
        ###################
        # train the model #
        ###################
        model.train()
        for batch_idx, (data, target) in enumerate(loaders['train']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## find the loss and update the model parameters accordingly
            ## record the average training loss, using something like
            ## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            
            optimizer.zero_grad()
            
            output = model(data)
            
            loss = criterion(output, target)
            loss.backward()
            
            optimizer.step()
            
            train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            
        ######################    
        # validate the model #
        ######################
        model.eval()
        for batch_idx, (data, target) in enumerate(loaders['valid']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## update the average validation loss
            output = model(data)
            loss = criterion(output, target)
            valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))

            
        # print training/validation statistics 
        print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
            epoch, 
            train_loss,
            valid_loss
            ))
        
        ## TODO: save the model if validation loss has decreased
        if valid_loss_min > valid_loss:
            torch.save(model.state_dict(), save_path)
            valid_loss_min = valid_loss
            print("Saved Epoch ", epoch)
            
    # return trained model
    return model
In [ ]:
 
In [21]:
# train the model
ImageFile.LOAD_TRUNCATED_IMAGES = True
model_scratch = train(30, loaders_scratch, model_scratch, optimizer_scratch, 
                      criterion_scratch, use_cuda, 'model_scratch.pt')
ImageFile.LOAD_TRUNCATED_IMAGES = False 
Epoch: 1 	Training Loss: 5.521908 	Validation Loss: 5.498691
Saved Epoch  1
Epoch: 2 	Training Loss: 5.500995 	Validation Loss: 5.478280
Saved Epoch  2
Epoch: 3 	Training Loss: 5.490738 	Validation Loss: 5.450332
Saved Epoch  3
Epoch: 4 	Training Loss: 5.482618 	Validation Loss: 5.434368
Saved Epoch  4
Epoch: 5 	Training Loss: 5.462017 	Validation Loss: 5.424444
Saved Epoch  5
Epoch: 6 	Training Loss: 5.425567 	Validation Loss: 5.388500
Saved Epoch  6
Epoch: 7 	Training Loss: 5.417952 	Validation Loss: 5.376161
Saved Epoch  7
Epoch: 8 	Training Loss: 5.391443 	Validation Loss: 5.328588
Saved Epoch  8
Epoch: 9 	Training Loss: 5.388680 	Validation Loss: 5.326238
Saved Epoch  9
Epoch: 10 	Training Loss: 5.364281 	Validation Loss: 5.299261
Saved Epoch  10
Epoch: 11 	Training Loss: 5.348703 	Validation Loss: 5.293571
Saved Epoch  11
Epoch: 12 	Training Loss: 5.356635 	Validation Loss: 5.271445
Saved Epoch  12
Epoch: 13 	Training Loss: 5.330034 	Validation Loss: 5.248940
Saved Epoch  13
Epoch: 14 	Training Loss: 5.324604 	Validation Loss: 5.231279
Saved Epoch  14
Epoch: 15 	Training Loss: 5.299223 	Validation Loss: 5.231286
Epoch: 16 	Training Loss: 5.304514 	Validation Loss: 5.206268
Saved Epoch  16
Epoch: 17 	Training Loss: 5.262700 	Validation Loss: 5.188786
Saved Epoch  17
Epoch: 18 	Training Loss: 5.241982 	Validation Loss: 5.175612
Saved Epoch  18
Epoch: 19 	Training Loss: 5.245625 	Validation Loss: 5.137156
Saved Epoch  19
Epoch: 20 	Training Loss: 5.222218 	Validation Loss: 5.126354
Saved Epoch  20
Epoch: 21 	Training Loss: 5.188548 	Validation Loss: 5.121659
Saved Epoch  21
Epoch: 22 	Training Loss: 5.216404 	Validation Loss: 5.127140
Epoch: 23 	Training Loss: 5.189859 	Validation Loss: 5.112517
Saved Epoch  23
Epoch: 24 	Training Loss: 5.175354 	Validation Loss: 5.055776
Saved Epoch  24
Epoch: 25 	Training Loss: 5.173069 	Validation Loss: 5.104138
Epoch: 26 	Training Loss: 5.161029 	Validation Loss: 5.081158
Epoch: 27 	Training Loss: 5.139854 	Validation Loss: 5.048252
Saved Epoch  27
Epoch: 28 	Training Loss: 5.133308 	Validation Loss: 5.012115
Saved Epoch  28
Epoch: 29 	Training Loss: 5.151766 	Validation Loss: 5.060915
Epoch: 30 	Training Loss: 5.109743 	Validation Loss: 5.031248

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.

In [22]:
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))
In [23]:
def test(loaders, model, criterion, use_cuda):

    # monitor test loss and accuracy
    test_loss = 0.
    correct = 0.
    total = 0.

    model.eval()
    for batch_idx, (data, target) in enumerate(loaders['test']):
        # move to GPU
        if use_cuda:
            data, target = data.cuda(), target.cuda()
        # forward pass: compute predicted outputs by passing inputs to the model
        output = model(data)
        # calculate the loss
        loss = criterion(output, target)
        # update average test loss 
        test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
        # convert output probabilities to predicted class
        pred = output.data.max(1, keepdim=True)[1]
        # compare predictions to true label
        correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
        total += data.size(0)
            
    print('Test Loss: {:.6f}\n'.format(test_loss))

    print('\nTest Accuracy: %.3f%% (%2d/%2d)' % (
        100. * correct / total, correct, total))

# call test function    
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Test Loss: 5.019207


Test Accuracy: 10.048% (84/836)

Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)

You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).

If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.

In [24]:
## TODO: Specify data loaders
loaders_transfer = loaders_scratch.copy()

(IMPLEMENTATION) Model Architecture

Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.

In [25]:
import torchvision.models as models
import torch.nn as nn

## TODO: Specify model architecture 

model_transfer = models.vgg16(pretrained=True)

for p in model_transfer.parameters():
    p.requires_grad = False

classifier = nn.Sequential(
    nn.Linear(512 * 7 * 7, 512 * 8),
    nn.ReLU(),
    nn.Dropout(0.5),
    nn.Linear(512 * 8, 512),
    nn.ReLU(),
    nn.Dropout(0.5),
    nn.Linear(512, dog_klass_count)
)

model_transfer.classifier = classifier

if use_cuda:
    model_transfer = model_transfer.cuda()

Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.

Answer:

I borrowed th pretrained vgg16, which most propably catch enough of features needed to classify breed.

I added a classifier similar to one in model_scratch; fully connected with inputs from vgg15 feature layers, and output layer to classify the target 133 breeds

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.

In [26]:
import torch.optim as optim

criterion_transfer = criterion_transfer = nn.CrossEntropyLoss()

# only train the classifier! -> model_transfer.classifier.parameters()
optimizer_transfer = optim.Adam(model_transfer.classifier.parameters(), lr=0.001)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.

In [27]:
# train the model

n_epochs = 10
ImageFile.LOAD_TRUNCATED_IMAGES = True
model_transfer = train(n_epochs, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
ImageFile.LOAD_TRUNCATED_IMAGES = False
        
# load the model that got the best validation accuracy (uncomment the line below)
model_transfer.load_state_dict(torch.load('model_transfer.pt'))
Epoch: 1 	Training Loss: 4.638469 	Validation Loss: 3.093525
Saved Epoch  1
Epoch: 2 	Training Loss: 3.007122 	Validation Loss: 1.571494
Saved Epoch  2
Epoch: 3 	Training Loss: 2.150561 	Validation Loss: 1.123176
Saved Epoch  3
Epoch: 4 	Training Loss: 1.731069 	Validation Loss: 0.952284
Saved Epoch  4
Epoch: 5 	Training Loss: 1.512828 	Validation Loss: 0.891000
Saved Epoch  5
Epoch: 6 	Training Loss: 1.336071 	Validation Loss: 0.766122
Saved Epoch  6
Epoch: 7 	Training Loss: 1.241674 	Validation Loss: 0.781278
Epoch: 8 	Training Loss: 1.182016 	Validation Loss: 0.769760
Epoch: 9 	Training Loss: 1.174520 	Validation Loss: 0.714819
Saved Epoch  9
Epoch: 10 	Training Loss: 1.102989 	Validation Loss: 0.721784

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.

In [28]:
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
Test Loss: 0.786082


Test Accuracy: 76.794% (642/836)

(IMPLEMENTATION) Predict Dog Breed with the Model

Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.

In [29]:
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.

def predict_breed_transfer(img_path):
    # load the image and return the predicted breed
    img = preprocessed_image(img_path)
    
    model_transfer.eval()
    res = None
    with torch.no_grad():
        res = model_transfer(img)
        res = torch.argmax(res).item()
    model_transfer.train()
    
    return res

Step 5: Write your Algorithm

Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,

  • if a dog is detected in the image, return the predicted breed.
  • if a human is detected in the image, return the resembling dog breed.
  • if neither is detected in the image, provide output that indicates an error.

You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and human_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.

Some sample output for our algorithm is provided below, but feel free to design your own user experience!

Sample Human Output

(IMPLEMENTATION) Write your Algorithm

In [30]:
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.

def run_app(img_path):
    ## handle cases for a human face, dog, and neither
    
    if face_detector(img_path):
        print("human face detected, resembling dog breed class id is", predict_breed_transfer(img_path))
    elif dog_detector(img_path):
        print("dog detected,preditcted dog breed class id is", predict_breed_transfer(img_path))
    else:
        print("neither a human nor a dog detected")

        

Step 6: Test Your Algorithm

In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?

(IMPLEMENTATION) Test Your Algorithm on Sample Images!

Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.

Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.

Answer:

Use Augmented dataset

adding images that does not contain dogs - and of course labled as such - to datasets

making the network deeper; adding more conv layers

In [36]:
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.

from IPython.core import display
from PIL import Image, ImageFile

## suggested code, below

test_images = np.array(glob("test_images/*"))

for file in np.hstack((human_files[-3:], dog_files[-3:], test_images)):
    
#     try:
    print("processing "  + file)
    display.display(display.Image(file,width=200,height=200))
    img = cv2.imread(file)
    run_app(file)
processing /data/lfw/Ferenc_Madl/Ferenc_Madl_0002.jpg
human face detected, resembling dog breed class id is 131
processing /data/lfw/Jim_Flaherty/Jim_Flaherty_0001.jpg
human face detected, resembling dog breed class id is 131
processing /data/lfw/Stacey_Yamaguchi/Stacey_Yamaguchi_0001.jpg
human face detected, resembling dog breed class id is 48
processing /data/dog_images/valid/100.Lowchen/Lowchen_06682.jpg
dog detected,preditcted dog breed class id is 81
processing /data/dog_images/valid/100.Lowchen/Lowchen_06708.jpg
dog detected,preditcted dog breed class id is 100
processing /data/dog_images/valid/100.Lowchen/Lowchen_06684.jpg
dog detected,preditcted dog breed class id is 100
processing test_images/depositphotos_210005586-stock-photo-dog-trainer-siberian-husky-dog.jpg
human face detected, resembling dog breed class id is 42
processing test_images/Stray_dogs_crosswalk.jpg
human face detected, resembling dog breed class id is 120
processing test_images/boxer_dog_brown_face_pet_animal.jpg
dog detected,preditcted dog breed class id is 43
processing test_images/dog-with-human-body.jpg
neither a human nor a dog detected
processing test_images/puppyclass-10.jpg
dog detected,preditcted dog breed class id is 119
processing test_images/depositphotos_168749200-stock-photo-sportswoman-and-sportsman-with-dog.jpg
neither a human nor a dog detected
processing test_images/depositphotos_42928519-stock-photo-veterinarian-examining-dog.jpg
human face detected, resembling dog breed class id is 49
processing test_images/Askals_or_aspins_are_mongrel_dogs_in_the_Philippines.jpg
neither a human nor a dog detected
processing test_images/32683801096_5b7598b75c_b.jpg
human face detected, resembling dog breed class id is 77
In [ ]:
%%bash

jupyter nbconvert --to html dog_app.ipynb